(19)
(11) EP 3 845 918 B1

(12) EUROPEAN PATENT SPECIFICATION

(45) Mention of the grant of the patent:
21.08.2024 Bulletin 2024/34

(21) Application number: 21150354.5

(22) Date of filing: 06.01.2021
(51) International Patent Classification (IPC): 
G01R 31/367(2019.01)
G01R 31/392(2019.01)
G01R 31/3828(2019.01)
(52) Cooperative Patent Classification (CPC):
G01R 31/392; G01R 31/367; G01R 31/3828

(54)

METHOD AND SYSTEM FOR ONLINE ESTIMATION OF SOH AND RUL OF A BATTERY

VERFAHREN UND SYSTEM ZUR ONLINE-SCHÄTZUNG VON SOH UND RUL EINER BATTERIE

PROCÉDÉ ET SYSTÈME POUR L'ESTIMATION EN LIGNE DE L'ÉTAT DE SANTÉ ET DE LA DURÉE DE VIE UTILE RESTANTE D'UNE BATTERIE


(84) Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(30) Priority: 06.01.2020 IN 202021000556

(43) Date of publication of application:
07.07.2021 Bulletin 2021/27

(73) Proprietor: Tata Consultancy Services Limited
Maharashtra (IN)

(72) Inventors:
  • Desai, Saurabh Jaywant
    411013 Pune, Maharashtra (IN)
  • Agarwal, Shashank
    411013 Pune, Maharashtra (IN)
  • Runkana, Venkataramana
    411013 Pune, Maharashtra (IN)
  • Pareek, Aditya
    411013 Pune, Maharashtra (IN)
  • Ramanujam, Muralikrishnan
    411013 Pune, Maharashtra (IN)

(74) Representative: Goddar, Heinz J. 
Boehmert & Boehmert Anwaltspartnerschaft mbB Pettenkoferstrasse 22
80336 München
80336 München (DE)


(56) References cited: : 
WO-A1-2019/199219
US-A1- 2015 301 122
US-A1- 2018 143 257
US-B1- 10 371 753
US-A1- 2003 184 307
US-A1- 2016 209 472
US-A1- 2019 353 711
   
  • NUHIC ADNAN ET AL: "Health diagnosis and remaining useful life prognostics of lithium-ion batteries using data-driven methods", JOURNAL OF POWER SOURCES, ELSEVIER SA, CH, vol. 239, 13 December 2012 (2012-12-13), pages 680 - 688, XP028566893, ISSN: 0378-7753, DOI: 10.1016/J.JPOWSOUR.2012.11.146
  • BARRÉ ANTHONY ET AL: "A review on lithium-ion battery ageing mechanisms and estimations for automotive applications", JOURNAL OF POWER SOURCES, ELSEVIER SA, CH, vol. 241, 7 June 2013 (2013-06-07), pages 680 - 689, XP028675862, ISSN: 0378-7753, DOI: 10.1016/J.JPOWSOUR.2013.05.040
   
Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


Description

CROSS-REFERENCE TO RELATED APPLICATIONS AND PRIORITY



[0001] The present application claims priority from Indian provisional patent application no. 202021000556, filed on January 6, 2020.

TECHNICAL FIELD



[0002] The disclosure herein generally relates to battery management, and, more particularly, to a method and system for online battery management involving real-time estimation of State of Health (SOH) and Remaining Useful Life (RUL) of a battery, based on real-time data collected from the battery.

BACKGROUND



[0003] Battery is an integral part of all devices that are mobile. For example, devices such as but not limited to mobile phones, and cameras are equipped with rechargeable batteries. Performance and life span of such batteries deteriorates with time. A few examples of factors that affect performance and lifetime of batteries are cyclic life, temperature, recharge rate and so on. Battery management in this context refers to monitoring and assessing performance of a battery, and in turn estimating Remaining Useful Life (RUL) of the battery being monitored.

[0004] Many state of the art systems/methods exist for performing the battery management. However, process adopted, and type of parameters considered by each methods may vary. Also, one disadvantage of the state of the art methods and systems for the battery management is that they rely on static values collected as input, for the purpose of the RUL estimation.

[0005] Prior art RUL estimation methods using historical battery usage data and various forms of predictions and extrapolations, including the use of machine learning models, are known from US10371753, US2019/353711, WO2019/199219, US2018/143257, US2003/184307, US2016/209472, US2015/301122.

SUMMARY



[0006] The invention provides a method according to claim 1. There is also provided a corresponding system according to claim 2, and a computer readable medium according to claim 3.

[0007] Embodiments of the present disclosure present technological improvements as solutions to one or more of the above-mentioned technical problems recognized by the inventors in conventional systems. For example, in one embodiment, a processor implemented method for online battery management according to claim 1 is provided. In this method, among other steps, a real-time value of voltage and current of a battery being monitored are determined, via one or more hardware processors. Further a state of the battery is determined as one of charging, discharging, and rest, via the one or more hardware processors, based on the determined value of one of the current and voltage. Further, value of at least one of a cumulative charge (Qchar), a time elapsed (Telap), Operation Time (Topn), and Charging Time (Topn_ch) is determined based on the determined state of the battery, via the one or more hardware processors. Further, the determined value of the at least one of the Qchar, Telap, Topn , and Topn_ch is processed with a battery performance model, via the one or more hardware processors, and the processing involves determining correlation of the determined value of the at least one of the Qchar, Telap, Topn, and Topn_ch with a battery performance model, determining a State of Health (SOH) of the battery based on the determined correlation, and determining a Remaining Useful Life (RUL) of the battery based on the determined SOH of the battery.

[0008] In another aspect, a system for online battery management according to claim 2 is provided. The system includes, among other features and means, a memory storing instructions, one or more communication interfaces, and one or more hardware processors coupled to the memory via the one or more communication interfaces. The one or more hardware processors are configured by the instructions to determine real-time value of voltage and current of a battery being monitored. Further a state of the battery is determined as one of charging, discharging, and rest, via the one or more hardware processors, based on the determined value of one of the current and voltage. Further, value of at least one of a cumulative charge (Qchar), a time elapsed (Telap), Operation Time (Topn), and Charging Time (Topn_ch) is determined based on the determined state of the battery, via the one or more hardware processors. Further, the determined value of the at least one of the Qchar, Telap, and Topn, Topn_ch is processed with a battery performance model, via the one or more hardware processors, and the processing involves determining correlation of the determined value of the at least one of the Qchar, Telap, Topn, and Topn_ch with a battery performance model, determining a State of Health (SOH) of the battery based on the determined correlation, and determining a Remaining Useful Life (RUL) of the battery based on the determined SOH of the battery.

[0009] In yet another aspect, a non-transitory computer readable medium for battery management according to claim 3 is provided. The non-transitory computer readable medium includes, among other instructions, a plurality of instructions which when executed using one or more hardware processors, cause the one or more hardware processors to determine a real-time value of voltage and current of a battery being monitored. Further a state of the battery is determined as one of charging, discharging, and rest, via the one or more hardware processors, based on the determined value of one of the current and voltage. Further, value of at least one of a cumulative charge (Qchar), a time elapsed (Telap), Operation Time (Topn), and Charging Time (Topn_ch) is determined based on the determined state of the battery, via the one or more hardware processors. Further, the determined value of the at least one of the Qchar, Telap, and Topn, Topn_ch is processed with a battery performance model, via the one or more hardware processors, and the processing involves determining correlation of the determined value of the at least one of the Qchar, Telap, Topn, and Topn_ch with a battery performance model, determining a State of Health (SOH) of the battery based on the determined correlation, and determining a Remaining Useful Life (RUL) of the battery based on the determined SOH of the battery.

[0010] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.

BRIEF DESCRIPTION OF THE DRAWINGS



[0011] The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed principles:

FIG. 1 illustrates an exemplary system for battery monitoring, according to some embodiments of the present disclosure.

FIG. 2 is a flow diagram depicting steps involved in the process of determining State of Health (SOH) and Remaining Useful Life (RUL) of the battery, using the system of FIG. 1, according to some embodiments of the present disclosure.

FIG. 3 is a flow diagram depicting steps involved in the process of determining the RUL from a determined SOH of the battery, in accordance with some embodiments of the present disclosure.

FIG. 4 is an example flow diagram depicting calculation of various parameters for determining the SOH for different states of the battery, using the process in FIG. 2 and the system in FIG. 1, in accordance with some embodiments of the present disclosure.

FIGS. 5A through 5F are graphs depicting values of different parameters as estimated by the system of FIG. 1 during the battery monitoring and SOH and RUL estimation, using the process in FIG. 2, in accordance with some embodiments of the present disclosure.


DETAILED DESCRIPTION OF EMBODIMENTS



[0012] Exemplary embodiments are described with reference to the accompanying drawings. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. Wherever convenient, the same reference numbers are used throughout the drawings to refer to the same or like parts. While examples and features of disclosed principles are described herein, modifications, adaptations, and other implementations are possible without departing from the scope of the disclosed embodiments. It is intended that the following detailed description be considered as exemplary only, with the true scope being indicated by the following claims.

[0013] Referring now to the drawings, and more particularly to FIG. 1 through 5F, where similar reference characters denote corresponding features consistently throughout the figures, there are shown preferred embodiments and these embodiments are described in the context of the following exemplary system and/or method.

[0014] FIG. 1 illustrates an exemplary system for battery monitoring, according to some embodiments of the present disclosure. In an embodiment, the system 100 includes a processor (s) 104, communication interface device(s), alternatively referred as input/output (I/O) interface(s) 106, and one or more data storage devices or a memory 102 operatively coupled to the processor (s) 104. In an embodiment, the processor (s) 104, can be one or more hardware processors (104). In an embodiment, the one or more hardware processors (104) can be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Among other capabilities, the processor(s) 104 is configured to fetch and execute computer-readable instructions stored in the memory 102. In an embodiment, the system 100 can be implemented in a variety of computing systems, such as laptop computers, notebooks, hand-held devices, workstations, mainframe computers, servers, a network cloud and the like.

[0015] The I/O interface(s) 106 can include a variety of software and hardware interfaces, for example, a web interface, a Graphical User Interface (GUI), and the like and can facilitate multiple communications within a wide variety of networks N/W and protocol types, including wired networks, for example, LAN, cable, etc., and wireless networks, such as WLAN, cellular, or satellite. In an embodiment, the I/O interface (s) 106 can include one or more ports for connecting a number of devices to one another or to another server. For example, the I/O interface 106 enables the authorized user to access the system disclosed herein through the GUI and communicate with other similar systems 100.

[0016] The memory 102 may include any computer-readable medium known in the art including, for example, volatile memory, such as static random access memory (SRAM) and dynamic random access memory (DRAM), and/or non-volatile memory, such as read only memory (ROM), erasable programmable ROM, flash memories, hard disks, optical disks, and magnetic tapes. Thus, the memory 102 may comprise information pertaining to input(s)/output(s) of each step performed by the processor(s) 104 of the system 100 and methods of the present disclosure.

[0017] The system 100 performs the online battery management. As part of the online battery management, the system 100 performs State of Health (SOH) and Remaining Useful Life (RUL) estimation of the battery. Steps in the SOH and RUL estimation, as performed by the system 100, are explained in description of FIG. 2 through FIG. 4.

[0018] FIG. 2 is a flow diagram depicting steps involved in the process of determining SOH and RUL of the battery, using the system of FIG. 1, according to some embodiments of the present disclosure. The system 100 performs an online monitoring of one or more batteries so as to determine SOH and RUL of the one or more batteries. The term 'online estimation' refers to capability of the system 100 to remotely connect with the batteries to collect real-time value of one or more parameters to determine the SOH and RUL, and also refers to capability of the system 100 to dynamically determine the SOH and RUL of the battery based on dynamically collected values of various parameters of the battery.

[0019] The system 100 generates a battery performance model, which is a machine learning data model that is generated using training data including operational data of at least one battery, SOH and RUL of the battery determined over a period of time, values of various parameters corresponding to each of the determined SOH and RUL and so on. Such data are collectively referred to as 'historical data' pertaining to operation of the battery. Every time such historical data is collected, the system 100 uses one or more appropriate techniques to clean the collected data. Cleaning of the data includes, but not limited to, synchronization of datasets, outlier removal, and imputation of data. The cleaning of the collected data is performed to fine-tune the collected historical data to obtain 'cleaned operational data' which is stored in the memory 101 and which is used for further processing to determine SOH and RUL of the battery at any given point of time. From the cleaned operational data, the SOH, State of Charge (SOC), RUL and so on are extracted as Key Variable Indicators (KVIs).

[0020] Further, the system 100 uses one or more feature engineering mechanisms to extract additional features, which along with the KVIs form training data for generating the battery performance model. Any appropriate machine learning technique may be used by the system 100 to process the training data and to build the battery performance model. The battery performance model is stored in a database in the memory 102, and may be updated from time to time using new training data.

[0021] Process of determining SOH and RUL of a battery is explained below. For convenience, the SOH and RUL estimation is explained by considering one battery. However, it is to be noted that the system 100 may perform the SOH and RUL estimation for more than one battery at a time. Also, the SOH and RUL estimation performed by the system 100 is an 'online estimation'. The term 'online estimation' in this context refers to capability of the system 100 to remotely connect with the batteries to collect real-time value of one or more parameters to determine the SOH and RUL, and also refers to capability of the system 100 to dynamically determine the SOH and RUL of the battery based on dynamically collected values of various parameters of the battery.

[0022] The system 100 can be connected to the battery using appropriate means such as but not limited to wired or wireless means, for example, a voltmeter and an ammeter are used. The voltmeter and ammeter used can be part of the system 100 externally or may be stand-alone devices. When connected, the system 100 initially determines (202) value of voltage and current of the battery. Based on the determined value of the voltage and battery, the system 100 further determines (204) state of the battery as one of 'charging', 'discharging', and 'rest'. If the determined value of current is non-zero, then the system 100 considers the battery as in operations mode, and if the determined value of current is 'zero', then the system 100 determines the battery as in the state of 'rest'. If the battery in the operations mode is drawing current from an external power supply, then the system 100 determines that the battery is in the 'charging' state. If the charge is coming out of the battery, then the system 100 determines the battery as in 'discharging' state.

[0023] Further the system 100 determines (206) value of at least one of a plurality of parameters including cumulative charge (Qchar), time elapsed (Telap), Operation time (Topn), and Charging Time (Topn_ch). If the battery is in the discharging state, the system 100 checks whether the determined value of the voltage (V) exceeds a minimum threshold of voltage (Vmin). If V exceeds Vmin, then the system 100 determines values of Qchar, time elapsed Telap, and Topn as:







[0024] The charging time (Topn_ch) is one of a charging time if the battery is in charging state, discharging time if the battery is in discharging state, and elapsed time if the battery is in the state of rest. Based on state of the battery, one of the aforementioned three values is determined as:
  1. 1. For charging state:
    If the battery is in 'charging' state and if (V ≤ maximum threshold of voltage (Vmax)), then

  2. 2. For discharging state:
    If the battery is in discharging state and if V ≥ Vmin,

  3. 3. For battery in the state of rest:



[0025] Values of Vmin and Vmax may be manufacture-specified.

[0026] If V is less than Vmin, then the system 100 terminates the process of SOH and RUL estimation. If the battery is in the state of 'rest', then the system 100 determines value of only the time elapsed (Telap), as:



[0027] If the battery is in the 'charging' state, then the system 100 checks whether the determined value of the voltage (V) is less than a maximum threshold of voltage (Vmax) i.e. whether V < Vmax. If V is less than Vmax, then the system 100 determines values of Qchar, time elapsed Telap, and Topn as:







[0028] If V exceeds Vmax, then the system 100 terminates the process of SOH and RUL estimation. The type of parameters determined for each state of the battery and the equations (1) through (10) are depicted in FIG. 4. The equations for Topn_ch are not depicted in FIG. 4, however they are used when needed.

[0029] In the next step, the system 100 determines (208) correlation of the determined value of at least one of the Qchar, Telap, Topn, and Topn_ch with the battery performance model. The battery performance model has data pertaining to one or more SOH and RUL determined at one or more past instance of time and corresponding values of the Qchar, Telap, and Topn parameters, for the same battery and/or a plurality of batteries including the battery for which the SOH and RUL estimation is being performed. While determining the correlation, the system 100 searches for and identifies matching values of the one or more parameters being considered, and based on a match found, determines (210) the corresponding SOH as the SOH of the battery.

[0030] In the next step, the system 100 determines (212) the RUL of the battery. This process is depicted in FIG. 3. The system 100 determines the RUL based on a rate of deterioration of SOH of the battery over a period of time. The system 100 collects (302) historical information pertaining to at least one SOH determined for the battery being monitored, at a current instance of time (i.e. real-time online estimation) and over a plurality of past time instances. In an embodiment, multiple SOH values determined over a period of time are used by the system 100 for the RUL estimation. The system 100 subtracts the SOH determined (304) at step 210 (referred to as 'SOH determined at a future instance of time (step 306 in in FIG. 3') from the at least one SOH determined during at least one past instance of time. The system 100 then determines (308) difference between future instance of time to reach a SOH Threshold (which is predetermined by user/ manufacturer) for the first time and the at least one SOH at the current instance of time, and based on the determined difference, determines (310) the RUL of the battery. The system 100 may use a suitable approach/technique such as but not limited to a normal distribution approximation technique to determine the SOH and value of associated parameters at the future instance of time, and the determined values are used to further determine the RUL. The determined value of SOH and RUL are further used by the system 100 to update the battery performance model. This is depicted in FIG. 4.

Experimental Results:



[0031] 
Table. 1
Abso lute time (seco nds) Volt age Cur rent Profile (Chargin g (C)/Disc harging (D)/Rest (R)) Time (Rela tive in sec) Derived tags Features
Cumul ative elapse d time Cumul ative charge -in (Amp-Sec) Tot al cha rge tim e (se c) Elapse d Time Ratio Mean Cumul ative Charge Load
3000 3.2-4.2 V 2A C 3000 0 6000 300 0 0 2
4000 4.2-4.2 V 0 R 1000 1000 6000 300 0 0.25 2
5500 4.2-3.5 V 3A D 1500 1000 6000 300 0 0.1818 18182 2
5600 3.5-3.4 V 4A D 100 1000 6000 300 0 0.1785 71429 2
8100 3.4-4.2 V 2.5 A C 2500 1000 12550 550 0 0.1234 5679 2.2272 72727
1010 0 4.2-4.2 V 0 R 2000 3000 12550 550 0 0.2970 29703 2.2272 72727
1160 0 4.2-3.4 V 3A D 1000 3000 12550 550 0 0.2586 2069 2.2272 72727


[0032] Table. 1 shows results of method of extraction of features including Elapsed Time ratio and total or cumulative charge ratio in. It is to be noted that these two features are considered only for example purpose, however other appropriate parameters also can be used by the system 100. Cumulative nature of these two features makes implementation of the battery performance model easier and convenient. During a first point, a charging instance of 2A is considered, where voltage of the battery rose from 3.2 V to 4.2 V. This instance leads to a positive change of 6000Amp-sec in the total charge-in and elapsed time which accounts for resting period stood at 0 Sec. Second point refers to the resting period and hence leads to an elapsed time of 1000 sec with no change in cumulative charge-in. Third and fourth cycles refer to discharge by a constant current of 3 A and 4 A respectively, hence no change in both the features is witnessed. Fifth cycle refers to the charging by 2.5 A for a duration of 2500 Sec, which lead to a change in cumulative charge in to 12250. Sixth cycle again is another rest cycle which changes the elapsed time to 3000 Sec with no change in seventh cycle of discharging.

[0033] For online measurement, it may not be directly possible to calculate the elapsed time. A flag based mechanism is used by the system 100 to overcome this issue, and is explained below:
  1. a. Keep a check on current and voltage of the battery and compare instantaneous voltage of the battery.
  2. b. If, there is some charging or discharging happening in the battery, then Elapsed time remains at its previous value. Otherwise if the battery is at rest (not loaded or getting charged), elapsed time increases with resting time.
  3. c. Add Elapsed time with previous value, whenever charging and discharging commences.


[0034] Results of SOH prediction done by the system 100 is depicted in FIG. 5A.

[0035] Remaining useful Life (RUL) - of the battery refers to the capacity fade, where capacity degrades to less than a certain value (for example, 80%) of rated capacity of the battery. RUL requires prediction of SOH at future states i.e. at future instances of time. Most of the time, these predictions come with an upper and lower bound, which talks about the degree of uncertainty in the predicted value and the predictions are supposed to lie within this defined bound.

[0036] The system 100 uses a normal distribution approximation in this scenario. (as depicted in FIG. 5B), standard deviations (1.96 to be exact) account for 95% of the area coverage w.r.t. mean. Considering the same analogy, at time t_1, θ refers to the mean value of an engineered feature(elapsed time for ex.) w.r.t total cumulative time, then upper and lower bound of the feature are given by θ± 2σ, where σ refers to the standard deviation from the mean as shown in FIG. 5B.

[0037] A RUL algorithm used by the system 100 uses predicted value of state parameters to estimate state of health of the battery at any future instance of time. The algorithm works on the battery performance model, and predicts state of health of the battery based on estimated state variables value.
Table. 2
Elapsed Time Ratio Charge Time Ratio Absolute Time Cumulative Charge Time Mean Cumulative Load
0.14 0.25 4000000 1000000 1.9
0.15 0.25 4000000 1100000 1.85
0.16 0.245 4000000 1176000 1.84
0.22 0.2 600000 120000 1.7
0.23 0.2 6500000 1300000 1.6
  1. a. First, state variables are identified, value of which need to be predicted based on fixed values of other state variables. For example, in the case of mean cumulative charge and elapsed time ratio as state variables, the elapsed time ratio is predicted based on historical data for given value of cumulative charge. Mean cumulative charge load is also predicted based on ratio of charging time to absolute time, using the battery performance model. Likewise, other parameters also are calculated.
  2. b. Define values of state variables, as mentioned in the Table. 2. Here, every row in the table defines a specific time when SOH prediction needs to be made, based on estimated value of SOH at a given starting time with cumulative charge of 11000 units. Cumulative charge column gives the point of prediction for RUL, second column gives the estimated mean value of elapsed time when battery is cumulatively charged to the given number based on historical data and is given by θ.
  3. c. Third and fourth columns give upper and lower bound on mean value as θ ± 1.96σ respectively.


[0038] Further in next steps, the battery performance model is used to determine value of SOH at a future instance of time under certain defined usage conditions. The 'usage conditions' are assumptions that usage pattern of the battery being monitored is same as at least one usage pattern previously recorded i.e. a historical usage pattern, which is part of the historical data. Also, sometimes manufacturers of batteries may specify End of Life (EOL) of battery as 70% or 50% of a rated capacity. This information also is taken as input by the system 100 and is used for the RUL estimation). Graphs in FIG. 5C, show behavior of the battery performance model with progressing cycles. Though, with less cycle information, there can be some error in accurate predictions, but with every passing cycle the battery performance model self-learns and gives more accurate results.

[0039] RUL of the battery is predicted using above mentioned procedure for all the cycles. To estimate between any two cycles, the RUL is estimated at cycle n and cycle n-1 and then linearly interpolated in between as shown in FIG. 5D. FIG. 5E shows result of the RUL estimation performed by the system 100. (50% of data is used for training the battery performance model and the ratios and future parameters are calculated. The calculated values are used to predict the SOH of the battery at a future instance of time. As in FIG. 5E, end of actual capacity is considered as End of battery life. Error in RUL prediction for 50% training data was observed as 9 days.

[0040] The written description describes the subject matter herein to enable any person skilled in the art to make and use the embodiments. The scope of the subject matter embodiments is defined by the claims and may include other modifications that occur to those skilled in the art. Such other modifications are intended to be within the scope of the claims if they have similar elements that do not differ from the literal language of the claims or if they include equivalent elements with insubstantial differences from the literal language of the claims.

[0041] The embodiments of present disclosure herein addresses unresolved problem of online battery monitoring. The embodiment, thus provides a mechanism for determining a State of Health (SOH) of a battery using values of battery specific parameters collected dynamically/real-time. Moreover, the embodiments herein further provide a method and system for RUL estimation based on relative deterioration in the SOH of the battery over a period of time.

[0042] It is to be understood that the scope of the protection is extended to such a program and in addition to a computer-readable means having a message therein; such computer-readable storage means contain program-code means for implementation of one or more steps of the method, when the program runs on a server or mobile device or any suitable programmable device. The hardware device can be any kind of device which can be programmed including e.g. any kind of computer like a server or a personal computer, or the like, or any combination thereof. The device may also include means which could be e.g. hardware means like e.g. an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a combination of hardware and software means, e.g. an ASIC and an FPGA, or at least one microprocessor and at least one memory with software processing components located therein. Thus, the means can include both hardware means and software means. The method embodiments described herein could be implemented in hardware and software. The device may also include software means. Alternatively, the embodiments may be implemented on different hardware devices, e.g. using a plurality of CPUs.

[0043] The embodiments herein can comprise hardware and software elements. The embodiments that are implemented in software include but are not limited to, firmware, resident software, microcode, etc. The functions performed by various components described herein may be implemented in other components or combinations of other components. For the purposes of this description, a computer-usable or computer readable medium can be any apparatus that can comprise, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.

[0044] The illustrated steps are set out to explain the exemplary embodiments shown, and it should be anticipated that ongoing technological development will change the manner in which particular functions are performed. These examples are presented herein for purposes of illustration, and not limitation. Further, the boundaries of the functional building blocks have been arbitrarily defined herein for the convenience of the description. Alternative boundaries can be defined so long as the specified functions and relationships thereof are appropriately performed. Modifications will be apparent to persons skilled in the relevant art(s) based on the teachings contained herein. Such alternatives fall within the scope of the disclosed embodiments. Also, the words "comprising," "having," "containing," and "including," and other similar forms are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items. It must also be noted that as used herein and in the appended claims, the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise.

[0045] Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present disclosure. A computer-readable storage medium refers to any type of physical memory on which information or data readable by a processor may be stored. Thus, a computer-readable storage medium may store instructions for execution by one or more processors, including instructions for causing the processor(s) to perform steps or stages consistent with the embodiments described herein. The term "computer-readable medium" should be understood to include tangible items and exclude carrier waves and transient signals, i.e., be non-transitory. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, nonvolatile memory, hard drives, CD ROMs, DVDs, flash drives, disks, and any other known physical storage media.

[0046] It is intended that the disclosure and examples be considered as exemplary only, with a true scope of disclosed embodiments being indicated by the following claims.


Claims

1. A processor implemented method (200) for online battery management, comprising:

determining (202) real-time value of voltage and current of a battery being monitored, via one or more hardware processors;

determining (204) a state of the battery as at least one of charging, discharging, and rest, via the one or more hardware processors, based on the determined real-time value of at least one of the current and voltage;

determining (206) value of at least one of a cumulative charge (Qchar), a time elapsed (Telap), Operation time (Topn), and Charging Time (Topn_ch) based on the determined state of the battery, via the one or more hardware processors; and

processing the determined value of the at least one of the Qchar, Telap, Topn, and Topn_ch with a battery performance model to determine a State of Health (SOH) and Remaining Useful Life (RUL) of the battery, via the one or more hardware processors, comprising:

determining (208) a correlation of the determined value of the at least one of the Qchar, Telap, Topn, and Topn ch using the battery performance model, wherein the battery performance model is used to determine value of the SOH at a future instance of time under certain usage conditions, wherein the certain usage conditions are assumptions that a usage pattern of the battery being monitored is same as at least one usage pattern based on historical information, wherein the battery performance model is a self-learning machine learning model trained using training data comprising historical data, wherein the training data includes operational data of the battery, the SOH and the RUL of the battery determined over a period of time, values of various parameters corresponding to each of the determined SOH and the determined RUL, wherein the collected historical data is cleaned to obtain cleaned operational data to determine the SOH and the RUL of the battery at a given point of time, wherein the SOH of the battery, State of Charge (SOC) of the battery and the RUL of the battery are extracted as a plurality of Key Variables of Interest (KVI) using the cleaned operational data;

determining, in real-time, (210) the SOH of the battery based on the determined correlation; and

determining, in real-time, (212) the RUL of the battery based on a rate of deterioration of the determined SOH of the battery, wherein the step of determining the RUL of the battery based on the determined SOH of the battery comprises:

collecting (302) historical information, wherein the historical information comprises at least one SOH of the battery determined at a past time instance, and values of the cumulative charge (Qchar), the time elapsed (Telap), the Topn, and the Topn_ch corresponding to the determined at least one SOH;

determining (304) values of the cumulative charge (Qchar), the time elapsed (Telap), the Topn, and the Topn_ch for the future instance of time, based on the historical information, wherein the values of the Qchar, Telap, Topn, and Topn ch are determined if the determined state of the battery is at least one of charging and discharging, wherein the value of the Telap is determined if the determined state of the battery is 'rest';

determining (306) the SOH at the future instance of time, based on the determined values of the cumulative charge (Qchar), the time elapsed (Telap), the Topn, and the Topn_ch; and

processing the determined SOH and the determined values of Qchar, Telap, Topn, and the Topn_ch using the battery performance model, comprising:

comparing the SOH determined for the future instance of time with the at least one SOH determined at the past instance of time;

determining (308) difference between the SOH determined for the future instance of time and the at least one SOH determined at the past instance of time;

determining difference for the future instance of time to reach a SOH threshold for first time and the at least one SOH at current instance of time; and

determining (310) the RUL of the battery based on the determined difference between the SOH determined for the future instance of time and the at least one SOH determined at the past instance of time.


 
2. A system for online battery management, comprising:

a memory (102) storing instructions;

one or more communication interfaces (106); and

one or more hardware processors (104) coupled to the memory (102) via the one or more communication interfaces (106), wherein the one or more hardware processors (104) are configured by the instructions to:

determine real-time value of voltage and current of a battery being monitored;

determine a state of the battery as at least one of charging, discharging, and rest, based on the determined real-time value of at least one of the current and voltage;

determine value of at least one of a cumulative charge (Qchar), a time elapsed (Telap), Operation Time (Topn), and Charging Time (Topn_ch) based on the determined state of the battery; and

process the determined value of the at least one of the Qchar, Telap, Top,n, and Topn_ch with a battery performance model to determine a State of Health (SOH) and Remaining Useful Life (RUL) of the battery, the processing comprising:

determining correlation of the determined value of the at least one of the Qchar, Telap, and Topn with a battery performance model, wherein the battery performance model is used to determine value of the SOH at a future instance of time under certain usage conditions, wherein the certain usage conditions are assumptions that a usage pattern of the battery being monitored is same as at least one usage pattern based on historical information, wherein the battery performance model is a self-learning machine learning model trained using training data comprising historical data, wherein the training data includes operational data of the battery, the SOH and the RUL of the battery determined over a period of time, values of various parameters corresponding to each of the determined SOH and the determined RUL, wherein the collected historical data is cleaned to obtain cleaned operational data to determine the SOH and the RUL of the battery at a given point of time, wherein the SOH of the battery, State of Charge (SOC) of the battery and the RUL of the battery are extracted as a plurality of Key Variables of Interest (KVI) using the cleaned operational data;

determining, in real-time, the SOH of the battery based on the determined correlation; and

determining, in real-time, the Remaining Useful Life (RUL) of the battery based on a rate of deterioration of the determined SOH of the battery, wherein the system determines the RUL based on the determined SOH of the battery by:

collecting the historical information, wherein the historical information comprises at least one SOH of the battery determined at a past time instance, and values of the cumulative charge (Qchar), the time elapsed (Telap), the Topn, and the Topn_ch corresponding to the determined at least one SOH;

determining values of the cumulative charge (Qchar), the time elapsed (Telap), the Topn, and the Topn ch for the future instance of time, based on the historical information, wherein the values of the Qchar, Telap, Topn, and Topn ch are determined if the determined state of the battery is at least one of charging and discharging, wherein the value of the Telap is determined if the determined state of the battery is 'rest';

determining the SOH at the future instance of time, based on the determined values of the cumulative charge (Qchar), the time elapsed (Telap), the Topn, and the Topn_ch; and

processing the determined SOH and the determined values of Qchar, Telap, Topn, and the Topn_ch using the battery performance model, comprising:

comparing the SOH determined for the future instance of time with the at least one SOH determined at the past instance of time;

determining difference between the SOH determined for the future instance of time and the at least one SOH determined at the past instance of time;

determining difference for the future instance of time to reach a SOH threshold for first time and the at least one SOH at current instance of time; and

determining the RUL of the battery based on the determined difference between the SOH determined for the future instance of time and the at least one SOH determined at the past instance of time.


 
3. A non-transitory computer readable medium for battery management, wherein the non-transitory computer readable medium comprising a plurality of instructions which when executed using one or more hardware processors, cause the one or more hardware processors to perform the battery management by:

determining real-time value of voltage and current of a battery being monitored, via one or more hardware processors;

determining a state of the battery as at least one of charging, discharging, and rest, via the one or more hardware processors, based on the determined real-time value of at least one of the current and voltage;

determining value of at least one of a cumulative charge (Qchar), a time elapsed (Telap), Operation time (Topn), and Charging Time (Topn_ch) based on the determined state of the battery, via the one or more hardware processors; and

processing the determined value of the at least one of the Qchar, Telap, Topn, and Topn_ch with a battery performance model to determine a State of Health (SOH) and Remaining Useful Life (RUL) of the battery, via the one or more hardware processors, the processing comprising:

determining correlation of the determined value of the at least one of the Qchar, Telap, Topn, and Topn ch with the battery performance model, wherein the battery performance model is used to determine value of the SOH at a future instance of time under certain usage conditions, wherein the certain usage conditions are assumptions that a usage pattern of the battery being monitored is same as at least one usage pattern based on historical information, wherein the battery performance model is a self-learning machine learning model trained using training data comprising historical data, wherein the training data includes operational data of the battery, the SOH and the RUL of the battery determined over a period of time, values of various parameters corresponding to each of the determined SOH and the determined RUL, wherein the collected historical data is cleaned to obtain cleaned operational data to determine the SOH and the RUL of the battery at a given point of time, wherein the SOH of the battery, State of Charge (SOC) of the battery and the RUL of the battery are extracted as a plurality of Key Variables of Interest (KVI) using the cleaned operational data;

determining, in real-time, the SOH of the battery based on the determined correlation; and

determining, in real-time, the Remaining Useful Life (RUL) of the battery based on a rate of deterioration of the determined SOH of the battery, wherein the step of determining the RUL of the battery based on the determined SOH of the battery comprises:

collecting (302) the historical information, wherein the historical information comprises at least one SOH of the battery determined at a past time instance, and values of the cumulative charge (Qchar), the time elapsed (Telap), the Topn, and the Topn_ch corresponding to the determined at least one SOH;

determining (304) values of the cumulative charge (Qchar), the time elapsed (Telap), the Topn, and the Topn_ch for the future instance of time, based on the historical information, wherein the values of the Qchar, Telap, Topn, and Topn_ch are determined if the determined state of the battery is at least one of charging and discharging, wherein the value of the Telap is determined if the determined state of the battery is 'rest';

determining (306) the SOH at the future instance of time, based on the determined values of the cumulative charge (Qchar), the time elapsed (Telap), the Topn, and the Topn_ch; and

processing the determined SOH and the determined values of Qchar, Telap, Topn, and the Topn_ch using the battery performance model, comprising:

comparing the SOH determined for the future instance of time with the at least one SOH determined at the past instance of time;

determining (308) difference between the SOH determined for the future instance of time and the at least one SOH determined at the past instance of time;

determining difference for the future instance of time to reach a SOH threshold for first time and the at least one SOH at current instance of time; and

determining (310) the RUL of the battery based on the determined difference between the SOH determined for the future instance of time and the at least one SOH determined at the past instance of time.


 


Ansprüche

1. Prozessorimplementiertes Verfahren (200) zur Online-Batterieverwaltung, das Folgendes umfasst:

Bestimmen (202) eines Echtzeitwerts von Spannung und Strom einer Batterie, die überwacht wird, über einen oder mehrere Hardwareprozessoren;

Bestimmen (204) eines Zustands der Batterie als mindestens eines von Laden, Entladen und Ruhe über den einen oder die mehreren Hardwareprozessoren basierend auf dem bestimmten Echtzeitwert von mindestens einem von dem Strom und der Spannung;

Bestimmen (206) eines Werts von mindestens einem von einer kumulativen Ladung (Qchar), einer verstrichenen Zeit (Telap), Betriebszeit (Topn) und Ladezeit (Topn_ch) basierend auf dem bestimmten Zustand der Batterie über den einen oder die mehreren Hardwareprozessoren; und

Verarbeiten des bestimmten Werts des mindestens einen von Qchar, Telap, Topn und Topn_ch mit einem Batterieleistungsmodell, um einen Gesundheitszustand (SOH) und eine verbleibende Nutzungsdauer (RUL) der Batterie über den einen oder die mehreren Hardwareprozessoren zu bestimmen, das Folgendes umfasst:

Bestimmen (208) einer Korrelation des bestimmten Werts des mindestens einen von Qchar, Telap, Topn und Topn ch unter Verwendung des Batterieleistungsmodells, wobei das Batterieleistungsmodell verwendet wird, um einen Wert des SOH zu einem zukünftigen Zeitpunkt unter bestimmten Nutzungsbedingungen zu bestimmen, wobei die bestimmten Nutzungsbedingungen Annahmen sind, dass ein Nutzungsmuster der Batterie, die überwacht wird, das gleiche ist wie mindestens ein Nutzungsmuster basierend auf historischen Informationen, wobei das Batterieleistungsmodell ein selbstlernendes Maschinenlernmodell ist, das unter Verwendung von Trainingsdaten trainiert wird, die historische Daten umfassen, wobei die Trainingsdaten Betriebsdaten der Batterie, den SOH und die RUL der Batterie, die über einen Zeitraum bestimmt werden, Werte verschiedener Parameter, die jedem von dem bestimmten SOH und der bestimmten RUL entsprechen, beinhalten, wobei die gesammelten historischen Daten bereinigt werden, um bereinigte Betriebsdaten zu erhalten, um den SOH und die RUL der Batterie zu einem gegebenen Zeitpunkt zu bestimmen, wobei der SOH der Batterie, der Ladezustand (SOC) der Batterie und die RUL der Batterie als eine Mehrzahl von Schlüsselvariablen von Interesse (KVI) unter Verwendung der bereinigten Betriebsdaten extrahiert werden;

Bestimmen, in Echtzeit, (210) des SOH der Batterie basierend auf der bestimmten Korrelation; und

Bestimmen, in Echtzeit, (212) der RUL der Batterie basierend auf einer Verschlechterungsrate des bestimmten SOH der Batterie, wobei der Schritt des Bestimmens der RUL der Batterie basierend auf dem bestimmten SOH der Batterie Folgendes umfasst:

Sammeln (302) historischer Informationen, wobei die historischen Informationen mindestens einen SOH der Batterie, der zu einem vergangenen Zeitpunkt bestimmt wurde, und Werte der kumulativen Ladung (Qchar), der verstrichenen Zeit (Telap), des Topn und des Topn_ch, die dem bestimmten mindestens einen SOH entsprechen, umfassen;

Bestimmen (304) von Werten der kumulativen Ladung (Qchar), der verstrichenen Zeit (Telap), des Topn und des Topn_ch für den zukünftigen Zeitpunkt basierend auf den historischen Informationen, wobei die Werte der Qchar, Telap, Topn und Topn_ch bestimmt werden, wenn der bestimmte Zustand der Batterie mindestens eines von Laden und Entladen ist, wobei der Wert der Telap bestimmt wird, wenn der bestimmte Zustand der Batterie "Ruhe" ist;

Bestimmen (306) des SOH zu dem zukünftigen Zeitpunkt basierend auf den bestimmten Werten der kumulativen Ladung (Qchar), der verstrichenen Zeit (Telap), des Topn und des Topn_ch; und

Verarbeiten des bestimmten SOH und der bestimmten Werte von Qchar, Telap, Topn und des Topn_ch unter Verwendung des Batterieleistungsmodells, das Folgendes umfasst:

Vergleichen des SOH, der für den zukünftigen Zeitpunkt bestimmt wurde, mit dem mindestens einen SOH, der zu dem vergangenen Zeitpunkt bestimmt wurde;

Bestimmen (308) einer Differenz zwischen dem SOH, der für den zukünftigen Zeitpunkt bestimmt wurde, und dem mindestens einen SOH, der zu dem vergangenen Zeitpunkt bestimmt wurde;

Bestimmen einer Differenz für den zukünftigen Zeitpunkt, um einen SOH-Schwellenwert für die erste Zeit und den mindestens einen SOH zu dem aktuellen Zeitpunkt zu erreichen; und

Bestimmen (310) der RUL der Batterie basierend auf der bestimmten Differenz zwischen dem SOH, der für den zukünftigen Zeitpunkt bestimmt wurde, und dem mindestens einen SOH, der zu dem vergangenen Zeitpunkt bestimmt wurde.


 
2. System zur Online-Batterieverwaltung, das Folgendes umfasst:

einen Speicher (102), der Anweisungen speichert;

eine oder mehrere Kommunikationsschnittstellen (106); und

einen oder mehrere Hardwareprozessoren (104), die über die eine oder die mehreren Kommunikationsschnittstellen (106) mit dem Speicher (102) gekoppelt sind, wobei der eine oder die mehreren Hardwareprozessoren (104) durch die Anweisungen konfiguriert sind zum:

Bestimmen eines Echtzeitwerts von Spannung und Strom einer Batterie, die überwacht wird;

Bestimmen eines Zustands der Batterie als mindestens eines von Laden, Entladen und Ruhe basierend auf dem bestimmten Echtzeitwert von mindestens einem von dem Strom und der Spannung;

Bestimmen eines Werts von mindestens einem von einer kumulativen Ladung (Qchar), einer verstrichenen Zeit (Telap), Betriebszeit (Topn) und Ladezeit (Topn_ch) basierend auf dem bestimmten Zustand der Batterie; und

Verarbeiten des bestimmten Werts des mindestens einen von Qchar, Telap, Top,n und Topn_ch mit einem Batterieleistungsmodell, um einen Gesundheitszustand (SOH) und eine verbleibende Nutzungsdauer (RUL) der Batterie zu bestimmen, wobei das Verarbeiten Folgendes umfasst:

Bestimmen einer Korrelation des bestimmten Werts des mindestens einen von Qchar, Telap und Topn mit einem Batterieleistungsmodell, wobei das Batterieleistungsmodell verwendet wird, um einen Wert des SOH zu einem zukünftigen Zeitpunkt unter bestimmten Nutzungsbedingungen zu bestimmen, wobei die bestimmten Nutzungsbedingungen Annahmen sind, dass ein Nutzungsmuster der Batterie, die überwacht wird, das gleiche ist wie mindestens ein Nutzungsmuster basierend auf historischen Informationen, wobei das Batterieleistungsmodell ein selbstlernendes Maschinenlernmodell ist, das unter Verwendung von Trainingsdaten trainiert wird, die historische Daten umfassen, wobei die Trainingsdaten Betriebsdaten der Batterie, den SOH und die RUL der Batterie, die über einen Zeitraum bestimmt werden, Werte verschiedener Parameter, die jedem von dem bestimmten SOH und der bestimmten RUL entsprechen, beinhalten, wobei die gesammelten historischen Daten bereinigt werden, um bereinigte Betriebsdaten zu erhalten, um den SOH und die RUL der Batterie zu einem gegebenen Zeitpunkt zu bestimmen, wobei der SOH der Batterie, der Ladezustand (SOC) der Batterie und die RUL der Batterie als eine Mehrzahl von Schlüsselvariablen von Interesse (KVI) unter Verwendung der bereinigten Betriebsdaten extrahiert werden;

Bestimmen, in Echtzeit, des SOH der Batterie basierend auf der bestimmten Korrelation; und

Bestimmen, in Echtzeit, der verbleibenden Nutzungsdauer (RUL) der Batterie basierend auf einer Verschlechterungsrate des bestimmten SOH der Batterie, wobei das System die RUL basierend auf dem bestimmten SOH der Batterie durch Folgendes bestimmt:

Sammeln der historischen Informationen, wobei die historischen Informationen mindestens einen SOH der Batterie, der zu einem vergangenen Zeitpunkt bestimmt wurde, und Werte der kumulativen Ladung (Qchar), der verstrichenen Zeit (Telap), des Topn und des Topn_ch, die dem bestimmten mindestens einen SOH entsprechen, umfassen;

Bestimmen von Werten der kumulativen Ladung (Qchar), der verstrichenen Zeit (Telap), des Topn und des Topn_ch für den zukünftigen Zeitpunkt basierend auf den historischen Informationen, wobei die Werte der Qchar, Telap, Topn und Topn_ch bestimmt werden, wenn der bestimmte Zustand der Batterie mindestens eines von Laden und Entladen ist, wobei der Wert der Telap bestimmt wird, wenn der bestimmte Zustand der Batterie "Ruhe" ist;

Bestimmen des SOH zu dem zukünftigen Zeitpunkt basierend auf den bestimmten Werten der kumulativen Ladung (Qchar), der verstrichenen Zeit (Telap), des Topn und des Topn_ch; und

Verarbeiten des bestimmten SOH und der bestimmten Werte von Qchar, Telap, Topn und des Topn_ch unter Verwendung des Batterieleistungsmodells, das Folgendes umfasst:

Vergleichen des SOH, der für den zukünftigen Zeitpunkt bestimmt wurde, mit dem mindestens einen SOH, der zu dem vergangenen Zeitpunkt bestimmt wurde;

Bestimmen einer Differenz zwischen dem SOH, der für den zukünftigen Zeitpunkt bestimmt wurde, und dem mindestens einen SOH, der zu dem vergangenen Zeitpunkt bestimmt wurde;

Bestimmen einer Differenz für den zukünftigen Zeitpunkt, um einen SOH-Schwellenwert für die erste Zeit und den mindestens einen SOH zu dem aktuellen Zeitpunkt zu erreichen; und

Bestimmen der RUL der Batterie basierend auf der bestimmten Differenz zwischen dem SOH, der für den zukünftigen Zeitpunkt bestimmt wurde, und dem mindestens einen SOH, der zu dem vergangenen Zeitpunkt bestimmt wurde.


 
3. Nichtflüchtiges computerlesbares Medium zur Batterieverwaltung, wobei das nichtflüchtige computerlesbare Medium eine Mehrzahl von Anweisungen umfasst, die, wenn sie unter Verwendung eines oder mehrerer Hardwareprozessoren ausgeführt werden, den einen oder die mehreren Hardwareprozessoren veranlassen, die Batterieverwaltung durchzuführen durch:

Bestimmen eines Echtzeitwerts von Spannung und Strom einer Batterie, die überwacht wird, über einen oder mehrere Hardwareprozessoren;

Bestimmen eines Zustands der Batterie als mindestens eines von Laden, Entladen und Ruhe über den einen oder die mehreren Hardwareprozessoren basierend auf dem bestimmten Echtzeitwert von mindestens einem von dem Strom und der Spannung;

Bestimmen eines Werts von mindestens einem von einer kumulativen Ladung (Qchar), einer verstrichenen Zeit (Telap), Betriebszeit (Topn) und Ladezeit (Topn_ch) basierend auf dem bestimmten Zustand der Batterie über den einen oder die mehreren Hardwareprozessoren; und

Verarbeiten des bestimmten Werts des mindestens einen von Qchar, Telap, Topn und Topn_ch mit einem Batterieleistungsmodell, um einen Gesundheitszustand (SOH) und eine verbleibende Nutzungsdauer (RUL) der Batterie über den einen oder die mehreren Hardwareprozessoren zu bestimmen, wobei das Verarbeiten Folgendes umfasst:

Bestimmen einer Korrelation des bestimmten Werts des mindestens einen von Qchar, Telap, Topn und Topn_ch mit dem Batterieleistungsmodell, wobei das Batterieleistungsmodell verwendet wird, um einen Wert des SOH zu einem zukünftigen Zeitpunkt unter bestimmten Nutzungsbedingungen zu bestimmen, wobei die bestimmten Nutzungsbedingungen Annahmen sind, dass ein Nutzungsmuster der Batterie, die überwacht wird, das gleiche ist wie mindestens ein Nutzungsmuster basierend auf historischen Informationen, wobei das Batterieleistungsmodell ein selbstlernendes Maschinenlernmodell ist, das unter Verwendung von Trainingsdaten trainiert wird, die historische Daten umfassen, wobei die Trainingsdaten Betriebsdaten der Batterie, den SOH und die RUL der Batterie, die über einen Zeitraum bestimmt werden, Werte verschiedener Parameter, die jedem von dem bestimmten SOH und der bestimmten RUL entsprechen, beinhalten, wobei die gesammelten historischen Daten bereinigt werden, um bereinigte Betriebsdaten zu erhalten, um den SOH und die RUL der Batterie zu einem gegebenen Zeitpunkt zu bestimmen, wobei der SOH der Batterie, der Ladezustand (SOC) der Batterie und die RUL der Batterie als eine Mehrzahl von Schlüsselvariablen von Interesse (KVI) unter Verwendung der bereinigten Betriebsdaten extrahiert werden;

Bestimmen, in Echtzeit, des SOH der Batterie basierend auf der bestimmten Korrelation; und

Bestimmen, in Echtzeit, der verbleibenden Nutzungsdauer (RUL) der Batterie basierend auf einer Verschlechterungsrate des bestimmten SOH der Batterie, wobei der Schritt des Bestimmens der RUL der Batterie basierend auf dem bestimmten SOH der Batterie Folgendes umfasst:

Sammeln (302) der historischen Informationen, wobei die historischen Informationen mindestens einen SOH der Batterie, der zu einem vergangenen Zeitpunkt bestimmt wurde, und Werte der kumulativen Ladung (Qchar), der verstrichenen Zeit (Telap), des Topn und des Topn_ch, die dem bestimmten mindestens einen SOH entsprechen, umfassen;

Bestimmen (304) von Werten der kumulativen Ladung (Qchar), der verstrichenen Zeit (Telap), des Topn und des Topn_ch für den zukünftigen Zeitpunkt basierend auf den historischen Informationen, wobei die Werte der Qchar, Telap, Topn und Topn_ch bestimmt werden, wenn der bestimmte Zustand der Batterie mindestens eines von Laden und Entladen ist, wobei der Wert der Telap bestimmt wird, wenn der bestimmte Zustand der Batterie "Ruhe" ist;

Bestimmen (306) des SOH zu dem zukünftigen Zeitpunkt basierend auf den bestimmten Werten der kumulativen Ladung (Qchar), der verstrichenen Zeit (Telap), des Topn und des Topn_ch; und

Verarbeiten des bestimmten SOH und der bestimmten Werte von Qchar, Telap, Topn und des Topn_ch unter Verwendung des Batterieleistungsmodells, das Folgendes umfasst:

Vergleichen des SOH, der für den zukünftigen Zeitpunkt bestimmt wurde, mit dem mindestens einen SOH, der zu dem vergangenen Zeitpunkt bestimmt wurde;

Bestimmen (308) einer Differenz zwischen dem SOH, der für den zukünftigen Zeitpunkt bestimmt wurde, und dem mindestens einen SOH, der zu dem vergangenen Zeitpunkt bestimmt wurde;

Bestimmen einer Differenz für den zukünftigen Zeitpunkt, um einen SOH-Schwellenwert für die erste Zeit und den mindestens einen SOH zu dem aktuellen Zeitpunkt zu erreichen; und

Bestimmen (310) der RUL der Batterie basierend auf der bestimmten Differenz zwischen dem SOH, der für den zukünftigen Zeitpunkt bestimmt wurde, und dem mindestens einen SOH, der zu dem vergangenen Zeitpunkt bestimmt wurde.


 


Revendications

1. Procédé, mis en oeuvre par processeur (200), de gestion de batterie en ligne, comprenant le fait de :

déterminer (202) une valeur en temps réel d'une tension et d'un courant d'une batterie en cours de surveillance, par l'intermédiaire d'un ou plusieurs processeurs matériels ;

déterminer (204) un état de la batterie comme étant au moins l'un parmi un état de charge, un état de décharge et un état de repos, par l'intermédiaire dudit un ou desdits plusieurs processeurs matériels, sur la base de la valeur en temps réel déterminée d'au moins l'un(e) parmi le courant et la tension ;

déterminer (206) la valeur d'au moins l'un des éléments parmi une charge cumulée (Qchar), un temps écoulé (Telap), un temps de fonctionnement (Topn) et un temps de charge (Topn_ch), sur la base de l'état déterminé de la batterie, par l'intermédiaire dudit un ou desdits plusieurs processeurs matériels ; et

traiter la valeur déterminée dudit au moins un des éléments parmi Qchar, Telap, Topn et Topn_ch avec un modèle de performance de batterie pour déterminer un état de santé (SOH) et une durée de vie utile restante (RUL) de la batterie, par l'intermédiaire dudit un ou desdits plusieurs processeurs matériels, comprenant le fait de :

déterminer (208) une corrélation de la valeur déterminée dudit au moins un des éléments parmi Qchar, Telap, Topn et Topn_ch, à l'aide du modèle de performance de batterie, dans lequel le modèle de performance de batterie est utilisé pour déterminer une valeur de l'état SOH à une instance temporelle future dans des conditions d'utilisation données, dans lesquelles lesdites conditions d'utilisation données sont des hypothèses selon lesquelles un schéma d'utilisation de la batterie en cours de surveillance est identique à au moins un schéma d'utilisation basé sur des informations historiques, dans lequel le modèle de performance de batterie est un modèle d'apprentissage automatique auto-apprenant entraîné à l'aide de données d'entraînement comprenant des données historiques, dans lesquelles les données d'entraînement incluent des données opérationnelles de la batterie, l'état SOH et la durée de vie RUL de la batterie déterminés sur une période de temps, des valeurs de divers paramètres correspondant à chacun(e) de l'état SOH déterminé et de la durée de vie RUL déterminée, dans lesquelles les données historiques collectées sont nettoyées en vue d'obtenir des données opérationnelles nettoyées permettant de déterminer l'état SOH et la durée de vie RUL de la batterie à un instant donné, dans lequel l'état SOH de la batterie, l'état de charge (SOC) de la batterie et la durée de vie RUL de la batterie sont extraits sous la forme d'une pluralité de variables clés d'intérêt (KVI) à l'aide des données opérationnelles nettoyées ;

déterminer, en temps réel, (210) l'état SOH de la batterie sur la base de la corrélation déterminée ; et

déterminer, en temps réel, (212) la durée de vie RUL de la batterie sur la base d'un taux de détérioration de l'état SOH déterminé de la batterie, dans laquelle l'étape de détermination de la durée de vie RUL de la batterie sur la base de l'état SOH déterminé de la batterie comprend le fait de :

collecter (302) des informations historiques, dans lesquelles les informations historiques comprennent au moins un état SOH de la batterie déterminé à une instance temporelle passée, et des valeurs de la charge cumulée (Qchar), du temps écoulé (Telap), du temps de fonctionnement Topn, et du temps de charge Topn_ch correspondant audit au moins un état SOH déterminé ;

déterminer (304) des valeurs de la charge cumulée (Qchar), du temps écoulé (Telap), du temps de fonctionnement Topn, et du temps de charge Topn_ch pour l'instance temporelle future, sur la base des informations historiques, dans lesquelles les valeurs de Qchar, Telap, Topn et Topn_ch sont déterminées si l'état déterminé de la batterie est au moins l'un parmi l'état de charge et l'état de décharge, dans laquelle la valeur du temps écoulé Telap est déterminée si l'état déterminé de la batterie est l'état de « repos » ;

déterminer (306) l'état SOH à l'instance temporelle future, sur la base des valeurs déterminées de la charge cumulée (Qchar), du temps écoulé (Telap), du temps de fonctionnement Topn, et du temps de charge Topn_ch ; et

traiter l'état SOH déterminé et les valeurs déterminées de Qchar, Telap, Topn et Topn_ch, à l'aide du modèle de performance de batterie, comprenant le fait de :

comparer l'état SOH déterminé pour l'instance temporelle future audit au moins un état SOH déterminé à l'instance temporelle passée ;

déterminer (308) une différence entre l'état SOH déterminé pour l'instance temporelle future et ledit au moins un état SOH déterminé à l'instance temporelle passée ;

déterminer une différence pour que l'instance temporelle future atteigne un seuil d'état SOH pour la première fois et ledit au moins un état SOH à une instance temporelle en cours ; et

déterminer (310) la durée de vie RUL de la batterie sur la base de la différence déterminée entre l'état SOH déterminé pour l'instance temporelle future et ledit au moins un état SOH déterminé à l'instance temporelle passée.


 
2. Système pour une gestion de batterie en ligne, comprenant :

une mémoire (102) stockant des instructions ;

une ou plusieurs interfaces de communication (106) ; et

un ou plusieurs processeurs matériels (104) couplés à la mémoire (102) par l'intermédiaire de ladite une ou desdites plusieurs interfaces de communication (106), dans lequel ledit un ou lesdits plusieurs processeurs matériels (104) sont configurés, par les instructions, de manière à :

déterminer une valeur en temps réel d'une tension et d'un courant d'une batterie en cours de surveillance ;

déterminer un état de la batterie comme étant au moins l'un parmi un état de charge, un état de décharge et un état de repos, sur la base de la valeur en temps réel déterminée d'au moins l'un(e) parmi le courant et la tension ;

déterminer la valeur d'au moins l'un des éléments parmi une charge cumulée (Qchar), un temps écoulé (Telap), un temps de fonctionnement (Topn) et un temps de charge (Topn_ch), sur la base de l'état déterminé de la batterie ; et

traiter la valeur déterminée dudit au moins un des éléments parmi Qchar, Telap, Topn et Topn_ch avec un modèle de performance de batterie pour déterminer un état de santé (SOH) et une durée de vie utile restante (RUL) de la batterie, le traitement comprenant le fait de :

déterminer une corrélation de la valeur déterminée dudit au moins un des éléments parmi Qchar, Telap, Topn, et Topn_ch à l'aide du modèle de performance de batterie, dans lequel le modèle de performance de batterie est utilisé pour déterminer une valeur de l'état SOH à une instance temporelle future dans des conditions d'utilisation données, dans lesquelles lesdites conditions d'utilisation données sont des hypothèses selon lesquelles un schéma d'utilisation de la batterie en cours de surveillance est identique à au moins un schéma d'utilisation basé sur des informations historiques, dans lequel le modèle de performance de batterie est un modèle d'apprentissage automatique auto-apprenant entraîné à l'aide de données d'entraînement comprenant des données historiques, dans lesquelles les données d'entraînement incluent des données opérationnelles de la batterie, l'état SOH et la durée de vie RUL de la batterie déterminés sur une période de temps, des valeurs de divers paramètres correspondant à chacun(e) de l'état SOH déterminé et de la durée de vie RUL déterminée, dans lesquelles les données historiques collectées sont nettoyées en vue d'obtenir des données opérationnelles nettoyées permettant de déterminer l'état SOH et la durée de vie RUL de la batterie à un instant donné, dans lequel l'état SOH de la batterie, l'état de charge (SOC) de la batterie et la durée de vie RUL de la batterie sont extraits sous la forme d'une pluralité de variables clés d'intérêt (KVI) à l'aide des données opérationnelles nettoyées ;

déterminer, en temps réel, l'état SOH de la batterie sur la base de la corrélation déterminée ; et

déterminer, en temps réel, la durée de vie RUL de la batterie sur la base d'un taux de détérioration de l'état SOH déterminé de la batterie, dans lequel le système détermine la durée de vie RUL de la batterie sur la base de l'état SOH déterminé de la batterie :

en collectant les informations historiques, dans lesquelles les informations historiques comprennent au moins un état SOH de la batterie déterminé à une instance temporelle passée, et des valeurs de la charge cumulée (Qchar), du temps écoulé (Telap), du temps de fonctionnement Topn, et du temps de charge Topn_ch correspondant audit au moins un état SOH déterminé ;

en déterminant des valeurs de la charge cumulée (Qchar), du temps écoulé (Telap), du temps de fonctionnement Topn, et du temps de charge Topn_ch pour l'instance temporelle future, sur la base des informations historiques, dans lesquelles les valeurs de Qchar, Telap, Topn et Topn_ch sont déterminées si l'état déterminé de la batterie est au moins l'un parmi l'état de charge et l'état de décharge, dans laquelle la valeur du temps écoulé Telap est déterminée si l'état déterminé de la batterie est l'état de « repos » ;

en déterminant l'état SOH à l'instance temporelle future, sur la base des valeurs déterminées de la charge cumulée (Qchar), du temps écoulé (Telap), du temps de fonctionnement Topn, et du temps de charge Topn_ch ; et

en traitant l'état SOH déterminé et les valeurs déterminées de Qchar, Telap, Topn et Topn_ch, à l'aide du modèle de performance de batterie, comprenant le fait de :

comparer l'état SOH déterminé pour l'instance temporelle future audit au moins un état SOH déterminé à l'instance temporelle passée ;

déterminer une différence entre l'état SOH déterminé pour l'instance temporelle future et ledit au moins un état SOH déterminé à l'instance temporelle passée ;

déterminer une différence pour que l'instance temporelle future atteigne un seuil d'état SOH pour la première fois et ledit au moins un état SOH à une instance temporelle en cours ; et

déterminer la durée de vie RUL de la batterie sur la base de la différence déterminée entre l'état SOH déterminé pour l'instance temporelle future et ledit au moins un état SOH déterminé à l'instance temporelle passée.


 
3. Support non transitoire lisible par ordinateur pour une gestion de batterie, dans lequel le support non transitoire lisible par ordinateur comprend une pluralité d'instructions qui, lorsqu'elles sont exécutées à l'aide d'un ou plusieurs processeurs matériels, amènent ledit un ou lesdits plusieurs processeurs matériels à mettre en oeuvre la gestion de batterie :

en déterminant une valeur en temps réel d'une tension et d'un courant d'une batterie en cours de surveillance, par l'intermédiaire d'un ou plusieurs processeurs matériels ;

en déterminant un état de la batterie comme étant au moins l'un parmi un état de charge, un état de décharge et un état de repos, par l'intermédiaire dudit un ou desdits plusieurs processeurs matériels, sur la base de la valeur en temps réel déterminée d'au moins l'un(e) parmi le courant et la tension ;

en déterminant la valeur d'au moins l'un des éléments parmi une charge cumulée (Qchar), un temps écoulé (Telap), un temps de fonctionnement (Topn) et un temps de charge (Topn_ch), sur la base de l'état déterminé de la batterie, par l'intermédiaire dudit un ou desdits plusieurs processeurs matériels ; et

en traitant la valeur déterminée dudit au moins un des éléments parmi Qchar, Telap, Topn et Topn_ch avec un modèle de performance de batterie pour déterminer un état de santé (SOH) et une durée de vie utile restante (RUL) de la batterie, par l'intermédiaire dudit un ou desdits plusieurs processeurs matériels, le traitement comprenant le fait de :

déterminer une corrélation de la valeur déterminée dudit au moins un des éléments parmi Qchar, Telap, Topn et Topn_ch, à l'aide du modèle de performance de batterie, dans lequel le modèle de performance de batterie est utilisé pour déterminer une valeur de l'état SOH à une instance temporelle future dans des conditions d'utilisation données, dans lesquelles lesdites conditions d'utilisation données sont des hypothèses selon lesquelles un schéma d'utilisation de la batterie en cours de surveillance est identique à au moins un schéma d'utilisation basé sur des informations historiques, dans lequel le modèle de performance de batterie est un modèle d'apprentissage automatique auto-apprenant entraîné à l'aide de données d'entraînement comprenant des données historiques, dans lesquelles les données d'entraînement incluent des données opérationnelles de la batterie, l'état SOH et la durée de vie RUL de la batterie déterminés sur une période de temps, des valeurs de divers paramètres correspondant à chacun(e) de l'état SOH déterminé et de la durée de vie RUL déterminée, dans lesquelles les données historiques collectées sont nettoyées en vue d'obtenir des données opérationnelles nettoyées permettant de déterminer l'état SOH et la durée de vie RUL de la batterie à un instant donné, dans lequel l'état SOH de la batterie, l'état de charge (SOC) de la batterie et la durée de vie RUL de la batterie sont extraits sous la forme d'une pluralité de variables clés d'intérêt (KVI) à l'aide des données opérationnelles nettoyées ;

déterminer, en temps réel, l'état SOH de la batterie sur la base de la corrélation déterminée ; et

déterminer, en temps réel, la durée de vie RUL de la batterie sur la base d'un taux de détérioration de l'état SOH déterminé de la batterie, dans laquelle l'étape de détermination de la durée de vie RUL de la batterie sur la base de l'état SOH déterminé de la batterie comprend le fait de :

collecter (302) les informations historiques, dans lesquelles les informations historiques comprennent au moins un état SOH de la batterie déterminé à une instance temporelle passée, et des valeurs de la charge cumulée (Qchar), du temps écoulé (Telap), du temps de fonctionnement Topn, et du temps de charge Topn_ch correspondant audit au moins un état SOH déterminé ;

déterminer (304) des valeurs de la charge cumulée (Qchar), du temps écoulé (Telap), du temps de fonctionnement Topn, et du temps de charge Topn_ch pour l'instance temporelle future, sur la base des informations historiques, dans lesquelles les valeurs de Qchar, Telap, Topn et Topn_ch sont déterminées si l'état déterminé de la batterie est au moins l'un parmi l'état de charge et l'état de décharge, dans laquelle la valeur du temps écoulé Telap est déterminée si l'état déterminé de la batterie est l'état de « repos » ;

déterminer (306) l'état SOH à l'instance temporelle future, sur la base des valeurs déterminées de la charge cumulée (Qchar), du temps écoulé (Telap), du temps de fonctionnement Topn, et du temps de charge Topn_ch ; et

traiter l'état SOH déterminé et les valeurs déterminées de Qchar, Telap, Topn et Topn_ch, à l'aide du modèle de performance de batterie, comprenant le fait de :

comparer l'état SOH déterminé pour l'instance temporelle future audit au moins un état SOH déterminé à l'instance temporelle passée ;

déterminer (308) une différence entre l'état SOH déterminé pour l'instance temporelle future et ledit au moins un état SOH déterminé à l'instance temporelle passée ;

déterminer une différence pour que l'instance temporelle future atteigne un seuil d'état SOH pour la première fois et ledit au moins un état SOH à une instance temporelle en cours ; et

déterminer (310) la durée de vie RUL de la batterie sur la base de la différence déterminée entre l'état SOH déterminé pour l'instance temporelle future et ledit au moins un état SOH déterminé à l'instance temporelle passée.


 




Drawing
































Cited references

REFERENCES CITED IN THE DESCRIPTION



This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

Patent documents cited in the description